Bacteremia is a severe infection of microorganisms in the bloodstream. The organisms cause a variety of symptoms including fever, shock, transient leukopenia and thrombocytopenia. Septic shock results when cardiac output cannot maintain the blood pressure due to a loss of intravascular volume. Septic shock can be caused by Gram-negative and Gram-positive bacteria, fungi, and although infrequently, by rickettsias or viruses. Escherichia coli account for a major portion of the cases of Gram-negative bacteremia.
The incidence of nosocomial bacteremia due to Gram-negative bacilli has been increasing since the 1950's despite the advent of antibiotics. It is now estimated to affect 260,000 to 300,000 patients per year in the United States with a mortality rate of 20 to 50% (1,2) even when treated under optimal situations. When the bacteremia is further complicated by renal or respiratory failure, the mortality approaches 90-100% (3).
Sepsis and septic shock may result from many causes, such as wound contamination in a traumatized patient, postoperative surgical complications, dissemination of a localized infection, or invasion of microorganisms through or around invasive instruments. Sepsis can be further complicated by immune suppression, as often occurs during critical illnesses. Most patients die from progressive multi-organ failure, which is thought to be due to the release of toxic substances from the infecting organisms (e.g., endotoxin), the release of secondary endogenous mediators, and altered metabolic status, resulting in progressive tissue ischemia.
Many of the symptoms and effects of Gram-negative bacteremia are consistent with the premise that endotoxin, or the lipopolysaccharide (LPS) component of the bacterial outer membrane, is the causative agent of the bacteria-induced shock. The Lipid-A portion of LPS is structurally conserved amongst the Enterobacteriaceae and is responsible for most of the biological to effects attributed to endotoxin. Although the length, number, saturation, and position of the acyl and acyloxyacyl chains are heterogeneous in the Lipid-A structure, the basic structure is common to all the Lipid-A molecules. Lipid-A molecules from some organisms such as E. coli are not as variable. LPS stimulates various cell types to release mediators, hormones or other factors, in particular tumor necrosis factor (TNF), interleukin-1 and interleukin-6, which in turn act on other organs or target tissues (4). Current therapeutic intervention includes antibiotics and fluids to increase intravascular volume, though these treatments are often unsuccessful at halting the cascading effects triggered by LPS. Antibiotics may reduce the bacteremia, but may also increase the amount of LPS shed into the bloodstream (5). As secondary complications occur, the physician must use other therapies to compensate for or to save the affected target organs. Experimental treatments employing murine and human monoclonal antibodies specific for Lipid-A are yielding encouraging results (6, 7).
It has also been proposed to employ antibodies specific for the Lipid-A moiety of LPS as treatment for septicemia or septic shock. Such antibodies have been IgM antibodies which are relatively large molecules and do not easily penetrate tissues.
Most monoclonal antibodies generated against the Lipid-A region of LPS have been produced by immunizing with killed cells of R-mutant gram-negative bacteria, or with such cells coated with additional Lipid-A or analogs. This approach has the disadvantage that the immune system is presented with natural Lipid-A structures at the same time as it is presented with the analog so that it is not certain exactly what the eliciting antigen for a given Mab might have been. In order to generate novel antibodies against Lipid-A with improved therapeutic properties including catalytic activity, it is desired to specifically stimulate the immune system with defined analogs.
Liposomes have been used widely and successfully as the basis for immunogens and vaccines to generate antibody responses to otherwise poorly immunogenic proteins or to obviate the need for harmful adjuvants (8, 9). It has also long been known that liposomes incorporating Lipid-A could induce antibodies capable of reacting with purified Lipid-A (10). Recently liposomes incorporating Lipid-A have been used to raise antibodies against short synthetic peptides which react with the native protein from which the peptide sequence was derived (11).
Recently it has been shown that monoclonal antibody fragments can be isolated by methods other than the conventional process of fusing specific B-cells with myeloma cells to generate hybridomas which secrete MAbs. The new method involves the isolation of the gene fragments encoding antibody molecules by their amplification, by the polymerase chain reaction (PCR), and cloning followed by expression as functional antigen binding molecules on the surface of filamentous phage particles (12, 13), or into the periplasmic space of bacteria infected with recombinant lambda-phage (14). The starting point for the PCR amplification of antibody-encoding gene fragments can be any of the following: splenocytes isolated from a mouse or other animal immunized with an antigen, e.g. a transition state analog; splenocytes isolated from an unimmunized animal; peripheral blood lymphocytes isolated from a human donor. Throughout this disclosure wherever reference is made to antibodies (or catalytic antibodies) or fragments thereof it is recognized that this applies both to antibodies derived by hybridoma technology and to antibodies isolated using the bacteriophage technologies outlined above. A fuller description of this technology is given in the copending application Ser. No. 07/841,648, filed Feb. 24, 1992, and incorporated herein by reference.
The manner in which catalytic antibodies carry out chemical reactions on substrates (or antigens) is essentially governed by the same theoretical principles that describe how enzymes carry out chemical reactions. For most chemical transformations to occur, substantial activation energy is required to overcome the energy barrier that exists between reactant and product. Enzymes catalyze chemical reactions by lowering the activation energy required to form the short-lived unstable chemical species found at the top of the energy barrier, known as the transition state (15, 16).
Four basic mechanisms are employed in enzymatic catalysis to lower the free energy of the rate limiting transition state, thereby accelerating the rate of a chemical reaction. Firstly, the active site of an enzyme is complementary in atomic and electronic structure to the transition state, such that the energy of the transition state is lower when bound to the enzyme than when free in solution. Secondly, general acid and base residues are often found optimally positioned for participation in catalysis within catalytic active sites causing the reaction to proceed via alternative and lower energy transition states. A third mechanism involves the formation of covalent enzyme-substrate intermediates. Fourth, model systems have shown that binding reactants in the proper orientation for reaction can increase the "effective concentration" of reactants by at least seven orders of magnitude (17).
Drawing upon this understanding of enzymatic catalysis, several antibodies with catalytic activity have been designed and isolated (18). Antibodies are elicited to compounds that resemble the transition state of a desired reaction (i.e., transition state analogs).
Several laboratories have studied the breakdown and detoxification of LPS and Lipid-A analogs (19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33). Their studies suggest that synthetic analogs of Lipid-A (E. coli) having one or more acyloxyacyl groups removed were less toxic than Lipid-A itself. The potential for detoxification of LPS is also demonstrated by the hydrolysis of the ester bonds of Lipid-A with the enzyme acyloxyacyl hydrolase, which removes the 3-hydroxytetradecanoyl chains of LPS, leaving hydroxyl groups (34, 35, 36, 37). Another study suggests that synthetic O-deacylated Lipid-A compounds are non-toxic (38, 39).
Based on these observations it is desired to develop transition state analogs of Lipid-A and LPS to elicit catalytic antibodies which can cleave the ester bonds of Lipid-A and LPS to detoxify the endotoxins. The target bonds for the catalytic antibodies are shown with arrows in FIG. 40 wherein the targets for ester bond hydrolysis are designated Immu-1, Immu-2, and Immu-3. It has also been shown that the synthetic monosaccharide analogs of Lipid-A and LPS are non-toxic (40). Based on this fact, it is also desired to design transition state analogs which will elicit catalytic antibodies which cleave the glycosidic bond of LPS and Lipid-A to the monosaccharide derivative (site Immu-4 in FIG. 40). It is desired to achieve this by eliciting catalytic antibodies to amidine transition state analogs of Lipid-A.
It is desired to employ catalytic antibodies to treat septicemia or septic shock. Catalytic antibodies are more efficacious than ordinary antibodies as a single molecule can inactivate many molecules of LPS. Catalytic antibodies can have better penetration of tissues as catalytic antibodies can be IgG-type antibodies or catalytically active fragments thereof (e.g. Fab, Fv, SCAb (single-chain antibodies) which are smaller in size). If these are conventional (non-catalytic) LPS binding antibodies of smaller size than IgM (e.g. IgG or fragments) they are unlikely to be effective in treatment or are less effective because they lack the effector functions associated with therapeutic action. On the contrary, the effector function of catalytic antibodies, namely the chemical detoxification of LPS, is mediated by the binding site such that any fragment which contains that site, however small, will be equally active. It is also desired to have new IgM-type antibodies which are also useful for treating septicemia or septic shock. Further, it is desired to be able to employ a cocktail of antibodies--of either various catalytic antibodies or of a mix of catalytic and binding antibodies (and of either IgG, IgM, antibody fragments, or a mix of IgG and IgM and antibody fragments)--to treat septicemia or septic shock. Thus, it is highly desired to obtain Lipid-A analogs which are capable of eliciting the desired catalytic and binding antibodies.
Further, therapeutic catalytic antibodies are desired for the treatment of sepsis and septic shock by means of cleavage of the glycosidic bond between the two monosaccharides for the detoxification of Lipid-A and LPS. In addition, antibodies generated against the analogs which bind to the Lipid-A moiety of LPS can have therapeutic efficacy in treatment of sepsis and are, therefore, also desired. Catalytic antibodies for reducing the toxicity of other compounds, e.g., glycosidic bond containing drugs or for activating compounds, e.g., glycosidic prodrugs, are also desired. It is also desired to be able to elicit such antibodies as an immunological response, for instance, to vaccinate against toxic substances. Thus, it is highly desired to obtain Lipid-A analogs which elicit these desired antibodies and which can be administered.
Lipid-A, which is responsible for endotoxic activity also exhibits beneficial antitumor (TNF inducing) and antiviral (IFN inducing) activities. Lipid-A analogs having decreased toxicity have been prepared and their biological activities studied (39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93). Monophosphorylated Lipid-A analogs exhibited Limulus, mitogenic, PBA, TNF (tumor necrosis factor)--inducing and IFN (interferon)--inducing activities, as well as protective activity against gram-negative bacteria, antiviral activity and antitumor activity.
It is also desired to prepare Lipid-A analogs having decreased toxicity but equivalent or increased beneficial antitumor and antiviral activities. The above-referenced studies demonstrate that both monosaccharide derivatives of Lipid-A having acyl chains and monophosphoryl Lipid-A analogs are non-toxic but retain some of their beneficial activities. The activity of these molecules may be dependent upon the composition of the lipid chains (e.g., the presence and position of functional groups and individual lipid chains); but, Applicants do not wish to necessarily be bound by any one particular theory. For instance, the presence of both a free hydroxyl group (i.e., hydroxy tetradecanoic) on the lipid chain attached to the 3-hydroxy group and a lauroyloxy tetradecanoic acid group attached to the 2-amino group of the monosaccharide was shown to produce a compound having good TNF inducing activity but only moderate IFN inducing activity. In contrast, reversing the position of these two groups, i.e., a lauroyloxy tetradecanoic acid group attached to the 3-hydroxy group and a hydroxy tetradecanoic acid group attached to the 2-amino position resulted in good IFN inducing activity but only moderate TNF inducing activity. These observations show that the presence and position of functional groups in the hydrophobic lipid region of Lipid-A and LPS can play an important functional role in the expression of activity.
It is desired to design new and novel compounds for use as therapeutic agents. For instance, it is desired to design new compounds having a hydroxy functionality in the lipid region without changing the hydrophobicity by eliminating an acyl chain.
It is especially desired to synthesize analogs having the pentavalent phosphorus in place of carbonyl carbon at the ester bond, allowing both a hydroxyl group and a lipid chain at the same position. The structure-activity relationship noted above indicates that such compounds have unique, improved and desired biological activities.
It is desired to obtain Lipid-A analogs which exhibit the beneficial activities of Lipid-A or of previous Lipid-A analogs with reduced toxicity or even without toxicity. It is also desired to obtain Lipid-A analogs having conformational rigidity (for instance so that the analog's activities are closer to the activities of Lipid-A), as well as a means for introducing nucleophilic functionality, such as hydroxyl, without decreasing lipophilicity (for instance by a pentavalent phosphorous in the lipophilic region). It is further desired to have a Lipid-A analog which allows direct attachment of pentavalent phosphorous to a sugar backbone. More particularly, it is desired to have both ester and hydroxy moieties at either the Immu-1 or Immu-2 positions (FIG. 40) so that the resultant Lipid-A analog exhibits superior IFN and TNF inducing activities.
When one further considers the four positions labeled in the accompanying structure of Lipid-A, e.g., Immu-1, Immu-2, Immu-3 and Immu-4 (FIG. 40), these would be sites for cleavage of Lipid-A by a catalytic antibody so as to detoxify Lipid-A. However, heretofore, no such catalytic antibody or compounds to elicit such have been described. It is thus desired to provide compounds which mimic the transition state (transition state analogs) of hydrolytic reactions at these positions in Lipid-A so as to elicit superior catalytic antibodies which can detoxify Lipid-A, e.g., by cleavage of ester bonds and/or glycosidic bonds of Lipid-A or LPS.